Suggested ideas for student’s PBL projects

Data-driven applications

Syntetic data

Possible projects/problems for a the thematic study of syntetic data could be chosen from this list:

1. Comparative study of open source platforms such as Gretel and Synthea for a specific datasets or datatype
2. Comparative study of ML model performance for generating synthetic data for various datatypes such as structured or unstructured
3. Evaluating the quality of data generated by defining metrics ( this can be further divided based on data types/formats)
4. Synthetic EHR data generation (where in FHIR frameworks comes into picture)
5. Replicating results of couple of papers and drawing out a comparative study
6. Applied projects like - DIPS Open, Epic Open, Cerner Open.

Contact: pavitra.chauhan@uit.no